Quality of institutional participation in multicenter clinical trials.

We studied the relation between an institution's degree of participation in a multicenter clinical trial and the quality of its participation. The quality of participation was judged on the basis of the percentage of patients entered into the trial who were in fact eligible, who were treated in accordance with the protocol, and for whom the appropriate forms were received. Those who satisfied these criteria were termed "valid" patients. Institutions were divided into two groups, "major participants" and "minor participants," according to the number of patients they entered in the trial. In a detailed analysis of a multicenter trial of chemotherapy for soft-tissue sarcoma, we found that major participants had a significantly higher percentage of valid patients than did minor participants (81 per cent vs. 39 per cent) and that the minor participants were actually detrimental to the study from both scientific and administrative viewpoints. These results suggest that institutions should not participate in multicenter studies unless they can enter some predetermined minimal number of patients per year.